<<

Environ. Sci. Technol. 2000, 34, 1297-1301

+ NpO2 cation, is highly mobile, biologically available, and Biological Reduction and Removal of very difficult to remove from solution by established methods. Np(V) by Two Microorganisms Prompted by early studies using uranium (6) and available biotechnology for radionuclide remediation (2) but previously constrained by the chemistry of neptunium(V) (4, 5), we JON R. LLOYD, PING YONG, AND describe a novel coupling of biological reduction of nep- LYNNE E. MACASKIE* tunium using Shewanella putrefaciens (previously docu- School of Biological Sciences, The University of Birmingham, mented to reduce uranium (VI) (7, 8) and technetium (VII) Edgbaston, Birmingham, B15 2TT U.K. (9)) to a -based biocrystallization method for uranium (10, 11), plutonium, and americium (11) which was ineffective alone against neptunium (V) (11). These studies were prompted by the observation that S. putrefaciens can The majority of the radionuclides generated by the use uranium as a terminal electron acceptor anaerobically, nuclear fuel cycle can be removed during established with concomitant reduction of U(VI) to U(IV) (12) and can remediation processes. However among the long-lived, also reduce Fe(III) to Fe(II) anaerobically (13). The redox 2+ R-emitting actinides neptunium(V) is recalcitrant to removal potentials for the appropriate couples are as follows: UO2 / 4+ ) + 4+ ) 3+ 2+ ) from solution by physicochemical or biotechnological U 0.32 V; NpO2 /Np 0.739 V; Fe /Fe 0.771 V methods. The latter include a biocrystallization process, (14); it is likely that Np(V) would be reduced similarly. based on the enzymatic liberation of phosphate as a Chemically, both ferrous ammonium sulfate and ascorbic precipitating ligand by a Citrobacter sp., which was previously acid can reduce Np(V) to the tetravalent state (14). It is likely shown to precipitate tetravalent actinides such as Th(IV) that Np(IV) could be precipitated as the phosphate; the of the tetravalent actinides are well documented and Pu(IV) as their corresponding phosphates. Np(V) was in the literature (4, 15), and previous studies have shown reduced to a lower valence (probably Np(IV)) by ascorbic that thorium(IV) phosphate can be removed by a phosphate- acid or biologically, using the reductive capability of liberating Citrobacter sp. (16). Actinide elements in the same Shewanella putrefaciens, but reduction alone did not oxidation state have very similar properties (17), and there desolubilize Np. However Np(V) was removed by the two are also many similarities in the chemical properties of the , S. putrefaciens and Citrobacter sp. in concert; trivalent actinide and lanthanide elements (18). bioreduction to Np(IV) by S. putrefaciens, together In a model system with La(III) and Th(IV) as “surrogates” with phosphate liberation by the Citrobacter sp., permitted for 241Am(III) and 239Pu(IV), and in tests using the corre- bioprecipitative removal of 237Np as well as its daughter sponding active nuclide species, metals were removed from 233protactinium. Tests were made possible by a novel solution as cell-bound phosphate biomineral, using pre- 2- technique permitting actinide separation by paper cipitant HPO4 ligand released via the activity of a cell-bound phosphatase of Citrobacter sp. (11). Immobilized Citrobacter chromatography followed by quantification of the radioactive > species using a phosphorImager. This study has implications cells in a column removed 90% of the input metal, to a load of9gofU/gofbiomass dry weight (19), as HUO2PO4 (10). for the development of methods to remove Np(V) from In contrast to this, and to the corresponding removal of La solution, by the simple combination of two biotechnological and Th phosphates (16, 20), Np(V) phosphate is fairly soluble methods, which can succeed where chemical treatments (4, 5), and Np was not removed by this method alone (11). are ineffective. Removal of Np by intercalation into pre-exisiting polycrys- talline uranyl or lanthanum phosphate is possible (21, 22), but such intercalation is restricted to Np(VI) (23), which arises Introduction in solution via disproportionation of Np(V), and the rate of this would be limited by the disproportionation rate of Np- In addition to the problem of extant radioactively contami- (V) in solution and the equilibrium position; in neutral nated sites, the stringent legislative requirements and current aqueous solution Np(V) predominates (24-26). concerns over the potential fate of “nuclear” wastes prompts Since the removal of Np(V) via both the biocrystallization evaluation and implementation of new technologies for the and intercalation routes is subject to chemical constraints, removal of contaminants from wastes and from the bio- we reasoned that if a biological system could reduce Np(V) sphere. The long-lived R-emitting actinides are problematic, to Np(IV), the latter could be removed via bioprecipitation, attributable to their very long half-lives and high radiotoxicity; as for Pu(IV) and Th(IV) (11, 27). Illustration of this formed these factors make their effective removal at the source both the purpose of the study. urgent and imperative. Among the transuranic nuclear fuel cycle elements Materials and Methods neptunium is problematic. It occurs in low-active wastes (1, 2) and, after uranium, can be the predominant radionuclide Microorganisms and Culture Conditions. Citrobacter sp. (10, (1, 2). It arises via decay of 241plutonium (half-life: 14.9 years) 11, 16, 27) was used by permission of Isis Innovation, Oxford, and 241americium (half-life: 433 years) to stable 237neptunium U.K. The was grown with vigorous aeration (30 °C) (half-life: 2.1 × 106 years). Eventually 237neptunium will in glycerol-glycerol 2-phosphate-based minimal medium, predominate in historical releases (2). While plutonium(IV) with biomass harvest and storage as previously described forms insoluble complexes in soils and sediments, limiting (16, 27, 28). Cells were suspended for use at a concentration environmental mobility and biological availability, neptuni- of 0.47 mg dry weight mL-1 in 40 mM MOPS-NaOH buffer um(V) complexes ligands poorly (3-5) and, as the R-emitting supplemented with 5 mM glycerol 2-phosphate (organic phosphate donor) and 100 mM ammonium acetate (pH 7) * Corresponding author phone: (44)-121-414-5889; fax: (44)-121- (to enhance the efficiency of actinide phosphate biopre- 414-5925; e-mail: [email protected]. cipitation (16, 28)).

10.1021/es990394y CCC: $19.00  2000 American Chemical Society VOL. 34, NO. 7, 2000 / ENVIRONMENTAL SCIENCE & TECHNOLOGY 9 1297 Published on Web 03/03/2000 Shewanella putrefaciens (ATCC 8071) was obtained from chromatography and visualized by emission of γ radiation Dr. D. R. Lovley (University of Massachusetts) and was grown using a phosphorimager-based technique (9). Air dried in the defined medium of Lovley and Phillips (29) under an chromatograms were wrapped in plastic laboratory film and anaerobic atmosphere consisting of N2 and CO2 (80:20). exposed to a storage phosphor screen (Molecular Dynamics, Sodium lactate (100 mM) and ferric citrate (50 mM) were Sevenoaks, Kent, U.K.). The spots of radioactivity were supplied as electron donor and acceptor, respectively. All visualized using a phosphorImager (Molecular Dynamics) 237 manipulations of S. putrefaciens were done under N2/CO2. after 48 h of exposure. In experiments where Np was used Prior to use, cells were collected from early stationary phase without 239Np γ tracer (the R-activity of 237Np does not cultures by centrfugation and washed three times in 40 mM penetrate the clingfilm used to wrap the samples and MOPS-NaOH as described by Lloyd and Macaskie (9). For therefore is not quantifiable by the phosphorImager tech- use in bioreduction tests cells were resuspended at a nique) the radionuclide was quantified directly in cell-free concentration of 0.31 mg mL-1. In some tests, cells were supernatants using a scintillation counter with R, â dis- washed three times and resuspended in carbonate buffer crimination. The chromatography paper, containing sepa- 237 233 (2.5 g NaHCO3, 1.5 g NaH2PO4, 0.1 g KCl per L distilled water) rated radionuclides ( Np and Pa daughter), was cut into and deaerated with 80%/20% N2/CO2. Where both organisms 1 cm sections, and each section was added to 10 mL “Ultima were used together (biomass content as above) the carrier Gold” scintillation cocktail (Packard Instrument Company, was 40 mM MOPS-NaOH with glycerol 2-phosphate and Meriden CT 06450, U.S.A.) and counted using a Tri-Carb ammonium acetate as above (pH 7). All cell suspensions (2 2700TR Liquid Scintillation Analyzer (Packard Instrument mL final volume) were contained in 12 mL serum bottles Company). Residual 237Np in solution was calculated by (Adelphi Tubes Ltd., Hayward’s Heath, Sussex, U.K.) at 30 adding 50 µL of aqueous sample to 10 mL of scintillation °C. cocktail. Uptake of 237Np by the cells (no attempt was made Chemicals and Radiochemicals. 237Np (containing daugh- to discriminate between precipitation, biosorption or bio- ter â-emitter 233Pa) used for most experiments was from AEA accumulation) was calculated by comparing disintegrations Fuel Services, Oxfordshire, U.K. 239Np (γ-emitter) was a gift min-1 in supernatants to disintegrations min-1 in cultures from BNFL (Salwick, Preston, U.K.) and was added to a carrier containing cells. 237Np and 233Pa were discriminated by solution of 237Np (233Pa free; purified by ion-chromatography counting in the R and â windows, respectively. by BNFL). The use of 239Np required specialist facilities at BNFL for its preparation and detection and it was difficult Results and Discussion to use routinely because of its short half-life (2.35 days). The Bioreduction tests utilized Shewanella putrefaciens, chosen 239/237 Np mix was used within 3 days of preparation, before on the basis of its ability to reduce U(VI) (7, 8) Fe(III) (13), 233 Pa was detectable. La, Th, and U nitrates were from BDH and Tc(VII) (9). The studies reported here were facilitated by (Poole, Dorset, U.K.), and other chemicals were from Sigma. the adaptation of a technique (“phosphorImager”) to visualize Reduction of Np(V) by Washed Cell Suspensions and â- and γ-species spatially separated chromatographically and Ascorbic Acid. 237Np (R-emitter containing daughter â-emit- imaged via a storage phosphor screen (9). Preliminary tests 233 233 ter Pa; concentration of Pa at use date was not used a surrogate metal mixture to establish likely Rf values determined) was added aseptically from an anaerobic stock for possible product valence species (III, IV, VI: Figure 1A) to a final concentration of 100 µM. 239Np (γ-emitter), when by direct staining of separated spots using arsenazo III. There used, was supplied to the cells at a concentration of 2 pM, is no convenient surrogate for Np(V) so tests were done using in a carrier solution of 2 µM 237Np, from which the 233Pa Np directly. A solution of 237Np, freshly prepared by ion- daughter element had been removed (above). Hydrogen was exchange chromatography and free from co-contaminants supplied in the headspace of the bottles when required as and daughter products (M. John: personal communication) an electron donor for Np reduction. Np(V) was reduced was donated by BNFL and gave no signal (Figure 1B) because chemically as required by the addition of ascorbic acid to a the phosphorimager technique does not detect R-activity final concentration of 1% (wt/vol) in cell-free control (above). A solution of 237Np obtained from a commercial experiments. source (i.e. containing 233Pa) gave no signal at the high-valence For bioreduction tests cells were incubated anaerobically positions (Figure 1C; cf.1A) but produced a nonmigrating (24 h) in the presence of either 100 µM 237Np or 2 µM 237Np/ radioactive spot, attributed to the active, â-emitting daughter 2pM 239Np as above. Cells and supernatant were separated 233protactinium(V). Alone among the pentavalent actinides, by centrifugation, and the residual species in the supernatant Pa(V) hydroxylates to give insoluble, colloidal hydroxide × (3 10 µL spots, dried under N2 between additions: total species (5, 31), distinguishing it from Np(V) which is highly load 30 µL) were separated chromatographically and quanti- soluble (4, 5). The insoluble 233Pa did not move from the fied as described below. origin in the chromatographic separation; therefore the Chromatographic Separation of Tri, Tetra- and Penta- behavior of 237Np could be followed in the presence of 233Pa. valent Actinides. Different oxidation states of the actinides To confirm the identity of Np in solution purified 237Np (no Np and Pa were separated chromatographically using What- other species were present and the R-activity is not detected: man 3MM paper (prewashed with 2 M HCl and several Figure 1B) was “spiked” with 239Np (â,γ-emitter), giving a changes of distilled water). The mobile phase was trimethyl- smeared streak by phosphorImager detection (Figure 1D), amine/acetone/formic acid 2:6:2 (v/v/v). The method was between the (IV) and (VI) positions identified in 1A and calibrated using “surrogate” elements La(III), Th(IV), and possibly corresponding to a redox equilibrium of Np(VI), U(VI) supplied as the nitrates (5 µg of each metal, 5 µLof Np(V), and Np(IV) but mainly Np(V). It is well-known from each applied to paper), separated as above, and visualized the literature that the Np(V) species is the most stable in on air-dried papers (30) by spraying with 0.15% arsenazo III solution. As a check the commercially obtained 237Np/233Pa 237 (w/v, aqueous (28, 30)). The Rf positions of each were was separated chromatographically (the R-emitting Np was determined by preliminary tests using one metal only, and not seen by the phosphorImager technique), and instead then the metals were tested in pairs and finally in a three cut sections were scintillation-counted with R,â-discrimina- metal mixture (30). The Rf position of each was the same tion. This showed clearly the R and â peaks of the respective throughout (30). Np/Pa components (Figure 1E). The short half-life of the Visualization and Quantification of 239Np, 237Np, and 239Np tracer (2.35 days) precluded its routine use; further 233Pa. 239Np in cell-free supernatants obtained after cen- tests used commercially obtained 237Np since good chro- trifugation (20 min at 13 000g) was separated using paper matographic separation was obtained between the parent

1298 9 ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 34, NO. 7, 2000 FIGURE 1. The use of microorganisms for the remediation of Np solutions. (A) Chromatographic separation of surrogate metal species in mixed valence solution. A mixture of La(III), Th(IV), and U(VI) (see Materials and Methods) was separated, and the Rf positions of each valence species were visualized using arsenazo III. (B,C) Chromatographic separation of species in solutions of 237Np (separated as in (A)). Detection (phosphorImager) was sensitive to â and γ-emissions via a storage phosphor screen (9). This confirms that the 237Np (r-emitter) is not detected by the technique and that the solution was purified of 233Pa (â-emitter) with testing as follows: (B) Chromatographic separation/phosphorImager detection of a solution (2 µM; 10 µL) of 237Np (r-emitter: not found by phosphorImager) separated from daughter products by ion exchange and supplied as purified solution by BNFL. (C) Chromatographic separation/phosphorImager detection of a commercially available solution of 237Np (100 µM) containing 237Np (r-emitting; phosphorImager “silent”) and â-emitting daughter 233Pa. 233 237 Details as in (B). The concentration was increased in order to obtain the Rf position of the Pa and to show that Np is not detectable by the technique. (D) Chromatographic separation and detection of Np species in a clean background of 237Np (shown in (B)) supplemented with 239Np (â,γ-emitter; 2 pM) and visualized using the phosphorImager following chromatographic separation as in (A). This shows the Rf position of the Np species in the starting solution. (E) Chromatographic separation and detection (by counting) of species within commercially obtained 237Np (r-emitter; 100 µM) and decay product 233Pa (â-emitter) separated chromatographically (as (A)) and detected 237 233 using scintillation counting. O: distribution of r-activity ( Np). Note similar Rf position to (D). b: distribution of â-activity ( Pa). Note 237 similar Rf position to that in (C). (F). Nuclide distribution following chemical reduction of commercially obtained Np using ascorbic acid (1% wt/vol aqueous final concentration), separated (as (A)) and counted (as (E)). Note that the Rf position has shifted from mid-way between the Th(IV) and U(VI) calibration positions ((E), cf. (A)) to the position of Th(IV) ((F), cf. (A)). Only 12% of the 237Np (100 µM) was removed from solution by centrifugation prior to analysis. (G). Hydrogen-dependent bioreduction of commercially available 237Np (mixture of 237Np and 233Pa: see (E)) with S. putrefaciens (see Materials and Methods). Note that biosorption of 233Pa was observed; this nuclide sorbs onto most surfaces (31). Little Np is apparent between the Rf positions of Th(IV) and U(IV) as shown in (A) and the Np position peaks at the 237 Rf position shown for Th(IV) in (A). Only 25% of the Np (100 µM) was removed with the biomass by centrifugation prior to analysis. (H). Bioreduction of 239Np in carbonate buffer. S. putrefaciens was prepared (as in (G)) and resuspended in 30 mM carbonate buffer (pH 7.5). The samples were supplemented with “cleaned” 237Np (bulk isotope: to 2 µM) and 239Np (spike: to 2 pM). 239Np species (â, γ-emission) were separated as in (A) and visualized as in (D). (I) Bioaccumulation of 233Pa and 237Np by Citrobacter sp. Chromatographic separation as in (A). Note loss of 233Pa from solution but negligible bioreduction or removal of Np. (J) Removal of 233Pa and 237Np by the concerted action of the two microorganisms. Suspensions of both organisms were under H2 as in (G) and were supplemented with 5 mM glycerol 2-phosphate and 100 mM ammonium acetate as in (I). Under these conditions, 95% of the 237Np was removed by centrifugation prior to analysis. nuclide and the daughter element. distinguishing the latter from Np (IV). Treatment of 237Np Chemical reduction of Np(V) using ascorbic acid gave a with Shewanella putrefaciens gave a similar species distribu- loss of activity in the Rf position between the Th(IV) and tion (1G) to that obtained using ascorbic acid (Figure 1F) U(VI) calibration positions and a migration to a lower Rf suggesting bioreduction of Np(V) to Np(IV). However in both value corresponding to that of Th(IV) (Figure 1F) and also cases the reduction was incomplete. This is probably because a new R-peak overlaying the â-peak of 233Pa, at the expense without removal of the reduced Np(IV) from solution 237 of Np(V) (1F). This suggested reduction of Np(V) to reestablishment of the disproportionation equilibrium was colloidal, nonmigrating hydroxylated species of Np(IV) via occurring in parallel to the reduction reactions. a reduced species. A portion of the Np did not correspond unequivocally to the position of Th(IV) or La(III) (cf. Figure The experiment was also done with 237/239Np, in carbonate 1A), but hydroxylation of the higher actinides and formation buffer and at a lower Np concentration to mimic more closely of colloidal material is more extensive than for Th(IV) (5), the situation of a natural water. Little â,γ activity was detected

VOL. 34, NO. 7, 2000 / ENVIRONMENTAL SCIENCE & TECHNOLOGY 9 1299 99 at the Rf positions corresponding to mobile valence (VI), (V), bioreduction of Tc(VII) by (36) and Des- 239 (IV), or (III) species; the recovered Np was at the position ulfovibrio desulfuricans (37) is now established, using H2 as corresponding to that of the ascorbate-reduced, nonmobile the electron donor. The potential harnessing of growth- 237Np (Figure 1H; cf. 1D). This suggests that in carbonate decoupled hydrogen-dependent metal reductase and phos- buffer and at low concentrations the soluble Np species are phatase activities of two organisms makes an integrated, removed from solution, but whether this was attributable to “clean” technique for bioremediation of neptunium wastes biosorption onto the biomass was not tested. uniquely possible. The use of hydrogen as the feedstock is In accordance with previous findings using Tc(VII) (9), noteworthy, since this introduces no additional organic load, little Np was removed from a 100 µM solution by S. and no waste product is generated from H2 oxidation. putrefaciens or following chemical reduction (25% and 12% Individually, Citrobacter, E. coli, and D. desulfuricans biore- removal of the input Np, respectively, corresponding to 85% actors have been used continuously over several weeks in and 88% of the Np remaining in solution), i.e., little was the removal of metal as phosphate (11, 19) and in the removed using bioreduction alone. This is in contrast to the bioreduction of 99Tc(VII) (38, 39), respectively, and there is bioreduction of U(VI), where U(IV) was precipitated as no reason why the coupled system should not be similarly insoluble UO2 (7, 8). Tetravalent actinides are insoluble at stable. Provision of H2 from a bottled supply could be pH 7 due to extensive hydroxylation and oxide formation (4, uneconomic or present a potential hazard. However in other 2+ o 5, 24). In contrast to the precipitation of UO2, the conversion tests we have shown that bioreduction of Pd to Pd by D. of hydroxylated Pu(IV) to the insoluble oxide (PuO2) is slow desulfuricans is supported by electrochemically generated (26, 32), i.e., Pu(IV) does not precipitate immediately, and it H2 which can be supplied on demand for continuous metal is likely that the hydroxylated neptunium(IV) is, similarly, a removal using immobilized cells in an electrobioreactor (40), soluble, nascent microcolloid which does not migrate chro- and such electrobioreactors have been shown to be stable matograpically and which does not precipitate onto biomass in operation over long periods for continuous metals recovery. per se, especially in the presence of 100 mM ammonium This preliminary study illustrates the potential for the acetate (above). Hence, the removal of Np from solution by biotechnological removal of Np(V), which is not easily ascorbic acid treatment or bioreduction alone is likely to be achieved chemically. Continuous-flow bioreactors for ac- low. These effects are difficult to confirm experimentally since tinide removal by Citrobacter sp. (11, 19) and metal reduction the speciation of Np in solution is complex (14) and the by E. coli and sulfate-reducing (37-39) are already physicochemical contribution of the bacterial cell surface established. It is anticipated that future studies will seek to (microenvironmental effects) is difficult to assess, given that establish the nature of the solution chemistry during Np(V) little is known of the reactive groups on the extracellular reduction, but these tests should be carried out in realistic polymeric matrix of the Shewanella sp. or, indeed, the way waste solutions, since these will contain other species which in which these might change according to the growth will influence the chemistry of the Np. As an example, many conditions of the organism. It can be concluded, however, wastes contain chelating agents (3, 41) and may also be at that S. purefaciens removed little Np from solution (Figure high or low pH, and a realistic assessment should be made 1G and see above). in an actual waste solution. It is hoped that this preliminary In separate experiments Citrobacter sp. did not remove demonstration will stimulate further study of the potential native (nonreduced) 237Np(V) (Figure 1I) in the presence of role of bioremediation in Np removal from real wastes in glycerol 2-phosphate, under which conditions Th(IV) and situ or by using “pump and treat” methods. Pu(IV) were removed via phosphate precipitation (11, 33). This is a simple hydrolysis reaction (cleavage of glycerol Acknowledgments 2-phosphate as a donor of inorganic phosphate at the cell We thank the U.K. Biotechnology and Biological Sciences surface); no input of energy is required, and the presence of Research Council (Grant No. GR/JO6276) and BNFL for hydrogen is not relevant to the phosphate pecipitation financial support. The phosphorImager was purchased with reaction. In the present system the Citrobacter sp. was shared equipment grants from the Wellcome Trust (Grant incorporated to scavenge specifically the Np(IV) in the No. 037160/Z/92) and the U.K. Medical Research Council presence of Np(V) by analogy with its abilty to remove Pu- (Grant No. G9216078MB). The storage phosphor screen was (IV) from solution as plutonium phosphate (11, 33). Although loaned from Molecular Dynamics Ltd. We thank Mr. S. Napier the removal of Np was negligible, 233Pa was removed (Figure and Mr. M. John of BNFL for the provision of laboratory 1I), in accordance with the high insolubility of protactinium facilities for work with 239Np and the gift of purified 237Np phosphate (5, 31). Without reduction Np(V) was retained at and 239Np and Drs. H. Eccles of BNFL and G. Zabierek of the the appropriate Rf position between that of Th(IV) and U(VI) Radiation Safety Office, University of Birmingham, for helpful in accordance with the inability of Citrobacter sp. to remove discussions. Np(V) per se (11) (see (Figure 1I). In the presence of S. putrefaciens 95% of the Np was removed from solution by Citrobacter sp. (Figure 1J), as anticipated from the low Literature Cited solubility products of the tetravalent actinide phosphates (1) Ashley, N. V.; Pope, N. R.; Roach, D. J. W. Department of the (4) and in accordance with previous studies using Th(IV) Environment: Commissioned Research on Radioactive Waste and Pu(IV) (11, 33). An identical result was obtained using Management; DOE/RW/88.008; 1987. Citrobacter sp. challenged with ascorbic acid-reduced Np, (2) Macaskie, L. E. Crit. Rev. Biotechnol. 1991, 11, 41. (3) Means, J. L,; Alexander, C. A. Nucl. Chem. Waste. Manage. 1981, suggesting strongly that the role of the Shewanella was to 2, 183. reduce Np(V) to Np(IV), in which form it becomes amenable (4) Allard., B. In Actinides in Perspective; Edelstein, N. M., Ed.; to the Citrobacter-mediated phosphate bioprecipitation. Pergamon: Oxford, U.K., 1982; pp 553-580. Thus, we have demonstrated the removal of Np(V) using (5) Ahrland, S.; Liljenzin, J. O.; Rydberg, J. In Comprehensive a coupled system, and the results suggest that this is via a Inorganic Chemistry. Vol. 5 The Actinides; Bailar, J. C., Emelius, concerted bioreduction and phosphate H. J., Nyholm, R., Trotman-Dickenson, A. F., Eds.; Pergamon Press: Oxford, U.K., 1973; pp 465-635. system. Previous attempts to accumulate Np(V) onto biomass (6) Woolfolk, C. A.; Whiteley, H. R. J. Bacteriol. 1962, 84, 647. gave only negligible uptake (11, 33-35), as anticipated from + (7) Lovley, D. R. Trends. Ecol. Evol. 1993,8,213. the chemistry of NpO2 . It was postulated previously (2, 6) (8) Lovley, D. R. Annu. Rev. Microbiol. 1993, 47, 263. that hydrogenase-mediated bioreduction of actinides would (9) Lloyd, J. R.; Macaskie, L. E. Appl. Environ. Microbiol. 1996, 62, be a distinct possibility. The role of hydrogenase in the 578.

1300 9 ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 34, NO. 7, 2000 (10) Macaskie, L. E.; Empson, R. M.; Cheetham, A. K.; Grey, C. P.; (27) Yong, P.; Macaskie, L. E. J. Chem. Technol. Biotechnol. 1998, 71, Skarnulis, A. J. Science 1992, 257, 782. 15. (11) Macaskie, L. E.; Jeong, B. C.; Tolley, M. R. FEMS Microbiol. Rev. (28) Yong, P.; Macaskie, L. E. J. Chem. Technol. Biotechnol. 1995, 63, 1994, 14, 351. 101. (12) Gorby, Y. A.; Lovley, D. R. Environ. Sci. Technol. 1992, 26, 205. (29) Lovley, D. R.; Phillips, E. J. P. Appl. Environ. Microbiol. 1988, (13) Obuekwe, C. D.; Westlake, D. W. S.; Cook, F. D. Can. J. Microbiol. 54, 1472. 1981, 27, 692. (14) Bailar, J. C.; Emelius, H. J.; Nyholm, R.; Trotman-Dickenson, A. (30) Yong, P. Ph.D. Thesis, University of Birmingham, U.K. 1996. F. Comprehensive Inorganic Chemistry. Vol. 5 Actinides; Per- (31) Kirby, H. W. In The Chemistry of the Actinide Elements; Katz, J. gamon Press: Oxford, U.K, 1973. J., Seaborg, G. T., Morss, L. R., Eds.; Chapman & Hall: London, - (15) Bamberger, C. E. In Handbook on the Physics and Chemistry of 1986; pp 102 168. the Actinides; Freeman, A. J., Keller, C., Eds.; Elsevier: Holland, (32) Ockenden, D. S.; Welch, G. A. J. Chem. Soc. 1956, 653, 3358. 1985; pp 289-303. (33) Yong, P.; Macaskie, L. E. J. Chem Technol. Biotechnol. 1998, 71, (16) Yong, P.; Macaskie, L. E. J. Chem. Technol. Biotechnol. 1995, 64, 15-26. 89. (34) Fisher, N. S.; Bjerrega, P.; Fowler, S. W. Limnol. Oceanogr. 1983, (17) Katz, J. J.; Seaborg, G. T.; Morss, L. R. Chemistry of the Actinide 28, 432. Elements; Chapman & Hall: London, 1986. (35) Strandberg, G.; Arnold, W. D. J. Ind. Microbiol. 1988, 3, 329. (18) Seaborg, G. T. Radiochim. Acta 1993, 61, 115. (19) Macaskie, L. E. J. Chem. Technol. Biotechnol. 1990, 49, 357. (36) Lloyd, J. R.; Cole, J. A.; Macaskie, L. E. J. Bacteriol. 1997, 179, (20) Tolley, M. R.; Strachan, L. F.; Macaskie, L. E. J. Ind. Microbiol 2014. 1995, 14, 271. (37) Lloyd, J. R.; Nolting, H. F.; Sole, V. A.; Bosecker, K.; Macaskie, (21) Tolley, M. R.; Macaskie, L. E. P. C. T. Patent Application 1994, L. E. Geomicrobiol. J. 1998, 15, 43. No. GB94/00626. (38) Lloyd, J. R.; Harding, C.; Macaskie, L. E. Biotechnol. Bioeng. (22) Bonthrone, K. M.; Basnakova G.; Lin, F.; Macaskie, L. E. Nature 1997, 55, 505. Biotechnol. 1996, 14, 635. (39) Lloyd, J. R., unpublished. (23) Dorhout, P. K.; Kissane, R. J.; Abney, K. D.; Avens, L. R.; Eller, (40) Yong, P., unpublished. P. G.; Ellis, A. B. Inorg. Chem. 1989, 28, 2926. (24) Allard, B.; Kipatsi, H.; Liljenzin, J. O. J. Inorg. Nucl. Chem. 1980, (41) Delegard, C. H. Radiochim. Acta 1987, 41, 11. 42, 1015. (25) Lieser, K. H.; Muhlenweg, U. Radiochim. Acta 1988, 43, 27. (26) Wildung, R. E.; Garland, T. E. In Transuranic Elements in the Received for review April 7, 1999. Revised manuscript re- Environment; DOE/NTIS, 22800; Hanson, W. C., Ed.; National ceived December 20, 1999. Accepted December 22, 1999. Technical Information Service, U.S. Department of Commerce: Springfield, VA, 1980. ES990394Y

VOL. 34, NO. 7, 2000 / ENVIRONMENTAL SCIENCE & TECHNOLOGY 9 1301